Criterion 1 The act must be aimed at attaining a political, economic, religious, or social goal. In terms of economic goals, the exclusive pursuit of profit does not satisfy this criterion. It must involve the pursuit of more profound, systemic economic change.

Criterion 2 There must be evidence of an intention to coerce, intimidate, or convey some other message to a larger audience (or audiences) than the immediate victims. It is the act taken as a totality that is considered, irrespective if every individual involved in carrying out the act was aware of this intention. As long as any of the planners or decision-makers behind the attack intended to coerce, intimidate or publicize, the intentionality criterion is met.

Criterion 3 The action must be outside the context of legitimate warfare activities. That is, the act must be outside the parameters permitted by international humanitarian law (particularly the prohibition against deliberately targeting civilians or non-combatants.

Doubt Terrorism Proper The existence of a “Yes” for “Doubt Terrorism Proper?” records reservation, in the eyes of GTD analysts, that the incident in question is truly terrorism. Such uncertainty, however, was not deemed to be sufficient to disqualify the incident from inclusion into the GTD. Furthermore, such a determination of doubt is subsequently coded by GTD analysts as conforming to one of four possible alternative designations: 1) Insurgency/Guerilla Action; 2) Internecine Conflict Action; 3) Mass Murder; or 4) Purely Criminal Act.

#Исследование базы данных о глобальном терроризме:

##Цели исследования: На заметку: сделать по график стран мира и числа терактов в них

Разобьём задачи на подзадачи: Интересующие подзадачи: Страны: Число преступлений в странах/их процентное соотношение Характер преступления в каждой стране Преступления: Характер и успешность преступления Зависимость преступления от страны Характер суицидальных атак Даты: Криминальные годы Продолжительные преступления Локация: Поквадратное исследование

Важное замечание: База данных образована таким образом, что некоторые столбцы пусты или неправильно заполнены вплоть до 1997-го года

library("tidyverse")
## Warning: пакет 'tidyverse' был собран под R версии 4.2.3
## Warning: пакет 'ggplot2' был собран под R версии 4.2.3
## Warning: пакет 'tibble' был собран под R версии 4.2.3
## Warning: пакет 'tidyr' был собран под R версии 4.2.3
## Warning: пакет 'readr' был собран под R версии 4.2.3
## Warning: пакет 'purrr' был собран под R версии 4.2.3
## Warning: пакет 'dplyr' был собран под R версии 4.2.3
## Warning: пакет 'stringr' был собран под R версии 4.2.3
## Warning: пакет 'forcats' был собран под R версии 4.2.3
## Warning: пакет 'lubridate' был собран под R версии 4.2.3
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.3     ✔ readr     2.1.4
## ✔ forcats   1.0.0     ✔ stringr   1.5.0
## ✔ ggplot2   3.4.4     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.0
## ✔ purrr     1.0.1     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library("ggplot2")
library("gganimate")
## Warning: пакет 'gganimate' был собран под R версии 4.2.3
library("rworldmap")
## Warning: пакет 'rworldmap' был собран под R версии 4.2.3
## Загрузка требуемого пакета: sp
## Warning: пакет 'sp' был собран под R версии 4.2.3
## ### Welcome to rworldmap ###
## For a short introduction type :   vignette('rworldmap')
library("leaflet")
## Warning: пакет 'leaflet' был собран под R версии 4.2.3
GT <- read.csv("globalterrorism.csv")

dateIssues.step Наблюдение 1: Не все даты соответствуют описанию из Codebook’а. Так eventid состоит из 12 цифр. Первые 8 — год (4 цифры), месяц (2 цифры)и день (2 цифры); последние 4 — порядковый номер для даты. При этом существуют несоответствия между датами и eventid и есть нулевой месяц и нулевой день (?) В целях экономии сил и времени будет пользоваться исходными данными.

Ошибочное наблюдение: в Codebook указано, что 0 - неопределённая дата

GT <- GT|>
  mutate(year = as.integer(str_sub(eventid, start=1, end=4)),
         month = as.integer(str_sub(eventid, start=5, end=6)),
         day = as.integer(str_sub(eventid, start=7,end=8)),
         .after=eventid)
GT <- GT|>
  mutate(date=as.Date(with(GT, paste(year, month, day, sep="-")), "%Y-%m-%d"),
         .after=day)|>
  mutate(idate=as.Date(with(GT, paste(iyear, imonth, iday, sep="-")), "%Y-%m-%d"),
         .after=iday)
dateValidator <- if_else(GT$date == GT$idate, T, F, NA)
length(dateValidator)
## [1] 181691
length(dateValidator[dateValidator == T])
## [1] 176727
length(dateValidator[dateValidator == F])
## [1] 5870
length(dateValidator[is.na(dateValidator)])
## [1] 906

Конец наблюдения 1.

Наблюдения 2: Даты

GTO <- GT|>
  select(!c(year, month, day, date))

Год

GTOY <- GTO|>
  group_by(iyear)|>
  summarize(yearNum = n())

rmarkdown::paged_table(GTOY|>arrange(desc(iyear)))
plot(GTOY$iyear, GTOY$yearNum, type="l",
     xlab = "Year", ylab = "Number of Crimes")

summary(GTOY$yearNum)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##     471    1364    2870    3866    4504   16903

Месяц

GTOY <- GTO|>
  group_by(imonth)|>
  summarize(monthNum = n())
barplot(height = GTOY$monthNum, 
        names.arg=GTOY$imonth,
        xlab = "Month",
        ylab = "Number of Crimes")

summary(GTOY$monthNum)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##      20   14180   15152   13976   15563   16875

День

GTOY <- GTO|>
  group_by(iday)|>
  summarize(dayNum = n())

barplot(GTOY$dayNum, 
        names.arg=GTOY$iday,
        xlab = "Day",
        ylab = "Number of Crimes")

summary(GTOY$dayNum)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##     891    5817    5933    5678    6027    6500

Длительность Для всех случаев, когда точная дата неизвестна, длительность события < 24 часов (одного дня)

Выводы: Присутствуют даты для которых extended = 1, но resolution = NA; Имеются данные с отрицательными временными промежутками (Ошибочные данные или запоздалое событие?) Самое длинное событие длилось 7324 дня (к сожалению, опечатка в базе данных)(индекс 4808) Событие:

GTO|>
  select(idate, extended)|>
  filter(is.na(idate) && extended == 1)
## Warning: There was 1 warning in `filter()`.
## ℹ In argument: `is.na(idate) && extended == 1`.
## Caused by warning in `is.na(idate) && extended == 1`:
## ! 'length(x) = 181691 > 1' перекодирован в 'logical(1)'
## [1] idate    extended
## <0 строк> (или 'row.names' нулевой длины)
GTD <- GTO|>
  select(eventid,idate, extended, resolution)|>
  group_by(idate)|>
  mutate(resNum=as.numeric(as.Date(resolution, format="%m/%d/%Y")),
         idateNum=as.numeric(idate))|>
  # mutate(resolution=ifelse(is.na(resolution),0,resolution),
  #        idateNum=ifelse(is.na(idate),0,idateNum))|>
  mutate(duration = ifelse(extended==0,0, resNum-idateNum))|>group_by(duration)
rmarkdown::paged_table(GTD|>summarize(durationNum=n()))
rmarkdown::paged_table(GTO[GTD$duration>=1000&!is.na(GTD$duration),])

Страны: Исходя из карты, можно увидеть, какие страны и в каком числе имеют теракты

library("rworldmap")
GTV <- GTO|>
  select(country_txt)|>
  group_by(country_txt)|>
  summarize(country_terror_num=n(), country_terror_per=100*n()/nrow(GTO))
spdf<-joinCountryData2Map(GTV, joinCode = "NAME", nameJoinColumn = "country_txt")
## 187 codes from your data successfully matched countries in the map
## 18 codes from your data failed to match with a country code in the map
## 56 codes from the map weren't represented in your data
mapCountryData(spdf, nameColumnToPlot="country_terror_num",catMethod="fixedWidth",
               missingCountryCol=gray(.8),colourPalette = "topo",mapRegion="world")

rmarkdown::paged_table(GTV|>arrange(desc(country_terror_num)))
barplot(GTV$country_terror_num, names.arg=GTV$country_txt)

Формы терроризма:

GTD <- GTO|>
  select(eventid,doubtterr, alternative_txt, attacktype1_txt, attacktype2_txt,
         attacktype3_txt)|>
  mutate(alternative_txt = ifelse(doubtterr==0, "Terrorism", alternative_txt))|>
  mutate(alternative_txt = ifelse(doubtterr==-9, "Not Included", alternative_txt))

GTDdterr <- GTD|>group_by(alternative_txt)|>
  summarize(terrtype_num = n(), terrtype_per = 100*n()/nrow(GTO))|>arrange(desc(terrtype_num))
rmarkdown::paged_table(GTDdterr)
barplot(GTDdterr$terrtype_num, names.arg = GTDdterr$alternative_txt)

attacktype:

GTD|>ungroup()|>group_by(alternative_txt, attacktype1_txt, attacktype2_txt, attacktype3_txt)|>
  summarize(total_num = n(), total_per=100*n()/nrow(GTD))|>as.data.frame()|>
  arrange(desc(total_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'alternative_txt', 'attacktype1_txt',
## 'attacktype2_txt'. You can override using the `.groups` argument.

region: В топ 5 узнать, какие теракты происходили

GTO|>
  group_by(country_txt, region_txt, provstate, city, latitude, longitude, specificity, vicinity)|>
  summarize(terr_num = n(), terr_per=100*n()/nrow(GTO))|>
  arrange(desc(terr_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'country_txt', 'region_txt', 'provstate',
## 'city', 'latitude', 'longitude', 'specificity'. You can override using the
## `.groups` argument.

multiple:

GTO|>select(multiple)|>group_by(multiple)|>
  summarize(mult_num=n(),mult_per=100*n()/nrow(GTO))|>as.data.frame()|>
  rmarkdown::paged_table()

success:

GTO|>select(success)|>group_by(success)|>
  summarize(suc_num = n(), suc_per = 100*n()/nrow(GTO))|>as.data.frame()|>
  rmarkdown::paged_table()

suicide:

GTO|>select(suicide)|>group_by(suicide)|>
  summarize(su_num = n(), su_per=100*n()/nrow(GTO))|>as.data.frame()|>
  rmarkdown::paged_table()

motive:

GTO|>group_by(motive)|>
  summarize(mot_num=n(), mot_per=100*n()/nrow(GTO))|>
  arrange(desc(mot_num))|>
  rmarkdown::paged_table()

targtype:

GTO|>group_by(targtype1_txt, targsubtype1_txt, targtype2_txt, targsubtype2_txt, 
              targtype3_txt, targsubtype3_txt)|>
  summarize(targ_num = n(), targ_per = 100*n()/nrow(GTO))|>
  arrange(desc(targ_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'targtype1_txt', 'targsubtype1_txt',
## 'targtype2_txt', 'targsubtype2_txt', 'targtype3_txt'. You can override using
## the `.groups` argument.

corp:

GTO|>group_by(corp1, corp2)|>
  summarize(corp_num = n(), corp_per=100*n()/nrow(GTO))|>
  arrange(desc(corp_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'corp1'. You can override using the
## `.groups` argument.

target/natlty:

GTO|>group_by(target1, natlty1_txt, target2, natlty2_txt, target3, natlty3_txt)|>
  summarize(targ_num=n(),targ_per=100*n()/nrow(GTO))|>
  arrange(desc(targ_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'target1', 'natlty1_txt', 'target2',
## 'natlty2_txt', 'target3'. You can override using the `.groups` argument.

perpetrator group:

GTO|>group_by(gname,gsubname,guncertain1, gname2,gsubname2,guncertain2,
              gname3,gsubname3,guncertain3)|>
  summarize(g_num=n(),g_per=100*n()/nrow(GTO))|>
  arrange(desc(g_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname', 'gsubname', 'guncertain1',
## 'gname2', 'gsubname2', 'guncertain2', 'gname3', 'gsubname3'. You can override
## using the `.groups` argument.

nperps:

GTO|>group_by(nperps)|>
  summarize(np_num=n(),np_per=100*n()/nrow(GTO))|>
  arrange(desc(np_num))|>
  rmarkdown::paged_table()

nperpcap:

GTO|>group_by(nperpcap)|>
  summarize(np_num=n(),np_per=100*n()/nrow(GTO))|>
  arrange(desc(nperpcap))|>
  rmarkdown::paged_table()

claims:

GTO|>group_by(claimed, claimmode_txt, claim2, claimmode2_txt,
              claim3, claimmode3_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'claimed', 'claimmode_txt', 'claim2',
## 'claimmode2_txt', 'claim3'. You can override using the `.groups` argument.

nkill:

temp <- GTO|>group_by(nkill)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))

rmarkdown::paged_table(temp)
plot(y=log10(temp$nkill), x=log10(temp$uni_num), type='l', ylab="Kill Count", xlab="Number of Incidents")

Общее число жертв:

sum(GTO$nkill, na.rm=T)
## [1] 411868
sum(GTO$nwound, na.rm=T)
## [1] 523869
sum(GTO$nkillter, na.rm=T)
## [1] 58291
sum(GTO$nwoundte,na.rm=T)
## [1] 12061
sum(GTO$nkillus, na.rm=T)
## [1] 5391
sum(GTO$nwoundus, na.rm=T)
## [1] 4556

nkillus:

GTO|>group_by(nkillus)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

nkillter:

GTO|>group_by(nkillter)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

nwound:

GTO|>group_by(nwound)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

nwoundus:

GTO|>group_by(nwoundus)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

nwoundte:

GTO|>group_by(nwoundte)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

propertydamage:

GTO|>group_by(property, propextent_txt, propcomment, propvalue)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'property', 'propextent_txt',
## 'propcomment'. You can override using the `.groups` argument.

nhostkid:

GTO|>group_by(ishostkid, nhostkid, nhours, ndays)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'ishostkid', 'nhostkid', 'nhours'. You can
## override using the `.groups` argument.

divert:

GTO|>group_by(divert, kidhijcountry)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'divert'. You can override using the
## `.groups` argument.

international:

GTO|>group_by(INT_LOG, INT_IDEO, INT_MISC)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'INT_LOG', 'INT_IDEO'. You can override
## using the `.groups` argument.

hostkidoutcome:

GTO|>group_by(hostkidoutcome_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

ransom:

GTO|>group_by(ransomamt, ransompaid)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'ransomamt'. You can override using the
## `.groups` argument.
GTO|>group_by(dbsource)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

location:

GTO|>group_by(location)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

weapdet:

GTO|>group_by(weaptype1_txt, weapsubtype1_txt, weaptype2_txt, weapsubtype2_txt,
              weaptype3_txt, weapsubtype3_txt, weaptype4_txt, weapsubtype4_txt,
              weapdetail)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'weaptype1_txt', 'weapsubtype1_txt',
## 'weaptype2_txt', 'weapsubtype2_txt', 'weaptype3_txt', 'weapsubtype3_txt',
## 'weaptype4_txt', 'weapsubtype4_txt'. You can override using the `.groups`
## argument.
range_coord <- function(param){
  return(paste(param-param%%10, param-param%%10+10,sep="-"))
}
GTD <- GTO|>
  group_by(latitude, longitude)|>
  mutate(latitude = sapply(latitude, range_coord),
         longitude = sapply(longitude, range_coord))|>
  summarize(block_num=n(),block_per=100*n()/nrow(GTO))|>
  arrange(desc(block_num))
## `summarise()` has grouped output by 'latitude'. You can override using the
## `.groups` argument.
rmarkdown::paged_table(GTD)
GTO1970 <- GTO|>filter(iyear==1970)
map <- leaflet(GTO1970)|>
  addTiles()|>
  addCircleMarkers(popup=GTO1970$summary, label=GTO1970$summary)
## Assuming "longitude" and "latitude" are longitude and latitude, respectively
## Warning in validateCoords(lng, lat, funcName): Data contains 8 rows with either
## missing or invalid lat/lon values and will be ignored
map
leaf <- leaflet() %>%
  addTiles() %>%
  addSimpleGraticule(interval=10)
leaf

specificity/vicinity (Про точность данных):

GTO|>group_by(specificity, vicinity)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'specificity'. You can override using the
## `.groups` argument.

weaptype:

GTO|>group_by(weaptype1_txt, weapsubtype1_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'weaptype1_txt'. You can override using the
## `.groups` argument.

targtype:

GTO|>group_by(targtype1_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
GTO|>group_by(target1)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

Complex Analysis 1:

GTO|>group_by(country_txt, attacktype1_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(country_txt, desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'country_txt'. You can override using the
## `.groups` argument.

total bombing/explosion in each country

GTO|>filter(attacktype1_txt=="Bombing/Explosion")|>group_by(country_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()

attacktype/success:

GTO|>group_by(attacktype1_txt, attacktype2_txt, attacktype3_txt, success)|>
  summarize(succ_num=n(), succ_per=100*n()/nrow(GTO))|>
  arrange(attacktype1_txt,attacktype2_txt,attacktype3_txt,desc(success))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'attacktype1_txt', 'attacktype2_txt',
## 'attacktype3_txt'. You can override using the `.groups` argument.

attacktypes:

GTO|>select(iyear, attacktype1_txt,attacktype2_txt,attacktype3_txt)|>
  gather(key="iyear", value="attacktype", -iyear)|>
  filter(attacktype!="")|>
  count(iyear, attacktype)|>as.data.frame()|>
  rmarkdown::paged_table()

targets

GTO|>select(iyear, target1, target2, target3)|>
  gather(key="iyear", value="target", -iyear)|>
  filter(target!="")|>
  count(iyear, target)|>as.data.frame()|>
  arrange(iyear, desc(n))|>
  rmarkdown::paged_table()

targtype

GTO|>select(iyear, targtype1_txt,targtype2_txt,targtype3_txt)|>
  gather(key="iyear", value="target", -iyear)|>
  filter(target!="")|>
  count(iyear, target)|>as.data.frame()|>
  arrange(iyear, desc(n))|>
  rmarkdown::paged_table()

year/attacktype:

library(ggplot2)
library(gganimate)
library(gapminder)
## Warning: пакет 'gapminder' был собран под R версии 4.2.3
library(gifski)
## Warning: пакет 'gifski' был собран под R версии 4.2.3
GTA <- GTO|>select(iyear, attacktype1_txt,attacktype2_txt,attacktype3_txt)|>
  gather(key="year", value="attacktype", -iyear)|>
  filter(attacktype!="")|>
  count(iyear, attacktype)|>as.data.frame()|>
  rename(num=n)|>
  group_by(iyear)|>
  mutate(rank=min_rank(-num)*1)|>
  filter(rank<=5)|>
  ungroup()

p <- ggplot(GTA, aes(rank, group = attacktype, 
                     fill = as.factor(attacktype), color = as.factor(attacktype))) +
  geom_tile(aes(y = num/2,
                height = num,
                width = 0.9), alpha = 0.8, color = NA) +

  # text in x-axis (requires clip = "off" in coord_*)
  # paste(country, " ")  is a hack to make pretty spacing, since hjust > 1 
  #   leads to weird artifacts in text spacing.
  geom_text(aes(y = 0, label = paste(attacktype, " ")), vjust = 0.2, hjust = 1) +

  coord_flip(clip = "off", expand = FALSE) +
  scale_y_continuous(labels = scales::comma) +
  scale_x_reverse() +
  guides(color = FALSE, fill = FALSE) +

  labs(title='{closest_state}', x = "", y = "Number of Attack Type") +
  theme(plot.title = element_text(hjust = 0, size = 22),
        axis.ticks.y = element_blank(),  # These relate to the axes post-flip
        axis.text.y  = element_blank(),  # These relate to the axes post-flip
        plot.margin = margin(1,1,1,4, "cm")) +

  transition_states(iyear, transition_length = 4, state_length = 1) +
  ease_aes('cubic-in-out')
## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
animate(p, 200, fps = 10, duration = 47, width = 1000, height = 800, renderer=gifski_renderer("attacktype.gif"))

Наблюдения по группировкам: число раненых/убитых

GTO|>group_by(gname)|>
  summarize(uni_num_kill=sum(nkill), uni_num_wound=sum(nwound),
            uni_num_kill_ter=sum(nkillter), uni_num_wound_ter=sum(nwoundte))|>
  arrange(desc(uni_num_kill))|>
  rmarkdown::paged_table()

Среднее местоположение терактов группировок. Знание региона разбивает действия группировок

GTO|>group_by(gname, region_txt)|>
  summarize(uni_long=mean(longitude),uni_lat=mean(latitude))|>
  arrange(gname)|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname'. You can override using the
## `.groups` argument.

Самый распространённый тип атаки у группировок:

GTO|>group_by(gname, attacktype1_txt)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  group_by(gname)|>
  summarize(frequent_method=attacktype1_txt[which.max(uni_num)])|>
  arrange(gname)|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname'. You can override using the
## `.groups` argument.

Самая распространённая цель у группировок:

GTO|>group_by(gname, target1)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  group_by(gname)|>
  summarize(frequent_target=target1[which.max(uni_num)])|>
  arrange(gname)|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname'. You can override using the
## `.groups` argument.
GTO|>group_by(gname, target1)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  arrange(desc(uni_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname'. You can override using the
## `.groups` argument.

Самые разнообразные цели у группировок:

GTO|>group_by(gname, target1, target2, target3)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  group_by(gname)|>
  summarize(target_num=n())|>
  arrange(desc(target_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname', 'target1', 'target2'. You can
## override using the `.groups` argument.

Сколько убитых/раненых каждым типом атаки:

GTO|>group_by(gname, target1)|>
  summarize(uni_num=n(),uni_per=100*n()/nrow(GTO))|>
  group_by(gname)|>
  summarize(frequent_target=target1[which.max(uni_num)])|>
  arrange(gname)|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'gname'. You can override using the
## `.groups` argument.

GIF со странами

GTA <- GTO|>group_by(iyear, country_txt)|>
  summarize(terr_num=n(),terr_per=100*n()/nrow(GTO))|>
  arrange(iyear, desc(terr_num))|>
  rmarkdown::paged_table()
## `summarise()` has grouped output by 'iyear'. You can override using the
## `.groups` argument.
library(ggplot2)
library(gganimate)
library(gapminder)
library(gifski)
theme_set(theme_classic())
GTA <- GTA|>
  group_by(iyear)|>
  mutate(rank = min_rank(-terr_num)*1)|>
  filter(rank<=10)|>
  ungroup()
p <- ggplot(GTA, aes(rank, group = country_txt, 
                     fill = as.factor(country_txt), color = as.factor(country_txt))) +
  geom_tile(aes(y = terr_num/2,
                height = terr_num,
                width = 0.9), alpha = 0.8, color = NA) +

  # text in x-axis (requires clip = "off" in coord_*)
  # paste(country, " ")  is a hack to make pretty spacing, since hjust > 1 
  #   leads to weird artifacts in text spacing.
  geom_text(aes(y = 0, label = paste(country_txt, " ")), vjust = 0.2, hjust = 1) +

  coord_flip(clip = "off", expand = FALSE) +
  scale_y_continuous(labels = scales::comma) +
  scale_x_reverse() +
  guides(color = FALSE, fill = FALSE) +

  labs(title='{closest_state}', x = "", y = "Number of Terrorist Acts") +
  theme(plot.title = element_text(hjust = 0, size = 22),
        axis.ticks.y = element_blank(),  # These relate to the axes post-flip
        axis.text.y  = element_blank(),  # These relate to the axes post-flip
        plot.margin = margin(1,1,1,4, "cm")) +

  transition_states(iyear, transition_length = 4, state_length = 1) +
  ease_aes('cubic-in-out')

animate(p, 200, fps = 10, duration = 47, width = 800, height = 600, renderer=gifski_renderer("terract.gif"))

individual:

GTO|>group_by(individual)|>
  summarize(num = n(), per = 100*n()/nrow(GTO))|>
  rmarkdown::paged_table()
GTO|>group_by(INT_LOG)|>
  summarize(num = n(), per = 100*n()/nrow(GTO))|>
  rmarkdown::paged_table()
GTO|>group_by(INT_IDEO)|>
  summarize(num = n(), per = 100*n()/nrow(GTO))|>
  rmarkdown::paged_table()
GTO|>group_by(INT_MISC)|>
  summarize(num = n(), per = 100*n()/nrow(GTO))|>
  rmarkdown::paged_table()

Число зависимость числа жертв от длительности теракта. Утверждение: чем дольше конфликт, тем больше жертв

GTO|>
  group_by(idate)|>
  mutate(resNum=as.numeric(as.Date(resolution, format="%m/%d/%Y")),
         idateNum=as.numeric(idate))|>
  # mutate(resolution=ifelse(is.na(resolution),0,resolution),
  #        idateNum=ifelse(is.na(idate),0,idateNum))|>
  mutate(duration = ifelse(extended==0,0, resNum-idateNum))|>group_by(duration)|>
  summarize(casualties = mean(nkill, na.rm=T)+mean(nwound, na.rm=T))|>
  arrange(desc(casualties))|>
  rmarkdown::paged_table()